To address the problems of low accuracy, difficult deployment and high calibration cost of visual manipulator in complex system environments, a robust joint modelling and optimization method for visual manipulators was proposed. Firstly, the subsystem models of the visual manipulator were integrated together, and the sample data such as servo motor rotation angles and manipulator end-effector coordinates were collected randomly in the workspace of the manipulator. Then, an Adaptive Multiple-Elites-guided Composite Differential Evolution algorithm with shift mechanism and Layered Optimization mechanism (AMECoDEs-LO) was proposed. Simultaneous optimization of the joint system parameters was completed by using the method of parameter identification. Principal Component Analysis (PCA) was performed by AMECoDEs-LO on stage data in the population, and with the idea of parameter dimensionality reduction, an implicit guidance for convergence accuracy and speed was realized. Experimental results show that under the cooperation of AMECoDEs-LO and the joint system model, the visual manipulator does not require additional instruments during calibration, achieving fast deployment and a 60% improvement in average accuracy compared to the conventional method. In the cases of broken manipulator linkages, reduced servo motor accuracy and increased camera positioning noise, the system still maintains high accuracy, which verifies the robustness of the proposed method.
Edge Computing (EC) and Simultaneous Wireless Information and Power Transfer (SWIPT) technologies can improve the performance of traditional networks, but they also increase the difficulty and complexity of system decision-making. The system decisions designed by optimization methods often have high computational complexity and are difficult to meet the real-time requirements of the system. Therefore, aiming at Wireless Sensor Network (WSN) assisted by EC and SWIPT, a mathematical model of system energy efficiency optimization was proposed by jointly considering beamforming, computing offloading and power control problems in the network. Then, concerning the non-convex and parameter coupling characteristics of this model, a joint optimization method based on deep reinforcement learning was proposed by designing information interchange process of the system. This method did not need to build an environmental model and adopted a reward function instead of the Critic network for action evaluation, which could reduce the difficulty of decision-making and improve the system real-time performance. Finally, based on the joint optimization method, an Improved Deep Deterministic Policy Gradient (IDDPG) algorithm was designed. Simulation comparisons were made with a variety of optimization algorithms and machine learning algorithms to verify the advantages of the joint optimization method in reducing the computational complexity and improving real-time performance of decision-making.
Aiming at the problem of malicious eavesdropping nodes in the energy limited multi-user Mobile Edge Computing (MEC) system, a joint Wireless Power Transfer (WPT) and MEC secure partial computing offloading programme was proposed. In order to minimize the energy consumption of the system Access Point (AP), the AP energy transmission covariance matrix, local CPU frequency, user unloading bits, user offloading time allocation and user transmission power were jointly optimized under the constraints of computing delay, secure offloading and energy capture. For the AP energy consumption minimization was a non-convex problem, firstly, the original non-convex problem was transformed into a convex problem by Difference of Convex Algorithm (DCA). Then, the optimal solution of the problem was obtained in semi-closed form by Lagrange duality method. When the number of computing tasks is 5 × 105 bits, compared with local computing offloading and secure full computing offloading, the energy consumption of secure partial offloading scheme was reduced by 61.3% and 84.4%, respectively; when the distance between eavesdropping nodes exceeds 25 m, the energy consumed by the secure partial offloading scheme is much less than those of local computing offloading and secure full computing offloading. The simulation results show that the proposed scheme can effectively reduce AP power consumption and enhance system performance gain while ensuring the secure offloading of the physical layer.
Considering the adoption of information and energy simultaneous transmission in wireless networks to improve the performance of wireless relay systems, a bidirectional transmission full-duplex relay system with self-energy recycling based on wireless radio frequency network was proposed by using Simultaneous Wireless Information and Power Transmission (SWIPT) technology. It is a new attempt to apply SWIPT in bidirectional full-duplex relay system. The energy-constrained destination node used the energy harvested from the relay and the loop channel to send feedback information, and the logical structure of the full-duplex relay system and the physical structure of the energy-constrained destination node were given. Then, the system performance was described by using the minimization of the system transmit power sum as the optimization target, the power allocation scheme was used for information decoding and energy harvest, the semi-definite programming, rank relaxation and Lagrange methods were used to transform the original non-convex optimization equation into a solvable convex optimization problem, and the solution of the problem was found. The relay transmission power, transmit beamforming vector and power allocation ratio were jointly optimized. Finally, the proposed system was compared with the traditional bidirectional transmission relay system by experimental simulator. The results verify that the self-energy recycling can not only eliminate self-interference, but also significantly optimize the system transmission power sum, and reveal that the proposed system has higher performance gain than the traditional bidirectional transmission system due to the combination of SWIPT technology and full-duplex relay system.
Scientists identify the species of whales based on the shape and the distinctive marks of the whale tails, but the process of recognition by human eyes and manual labeling is very cumbersome. The dataset of whale tail photo has the unbalanced data distribution, and some specific categories in the dataset have very few samples or even one sample. Besides, the samples have small individual differences and contain unknown categories, which leads to the difficulty in automatic labeling of whale identification by image classification. To solve the problem that metric learning is difficult to realize classification under this task, on the basis of Siamese Neural Network (SNN), the training batches were constructed dynamically by using Linear Assignment Problem (LAP) algorithm in the training process of hard-negative sample mining. Firstly, image feature vectors were extracted from the training samples, and the similarity metric of feature vector was calculated. Then, LAP was used to assign sample pairs to the model, training sample batches were constructed dynamically according to the metric score matrix, and the difficult sample pairs were targeted by trained. Experimental results on a whale tail image dataset with unbalanced data distribution and CUB 200-2001 dataset show that, the proposed algorithm can achieve good results in learning minority classes and classifying fine-grained images.
The result that current web search engineer returned were ranked mainly by their hyperlink analyse, not their content. To take the search results as an order collection, we used item frenqency statistic and calculated item position in every page by certain formula, by which we calculated each pages relativity and re-ranked the collection. The experiment results show that the pages which meet the users needs were concentrated ahead. In this way, The precision was enhanced. It can help user find information rapidly.